An Adaptive Low-latency Power Management Protocol for Wireless Sensor Networks

Energy conservation in wireless sensor networks is a critical issue. An efficient method to reduce power consumption consists in powering off the nodes' wireless transceiver when communication is not needed. Under this approach sleep/wakeup schedules of different nodes have to be synchronized. In addition, during the sleep phases nodes cannot communicate, and this might result in high delay. In this paper we introduce an adaptive and low latency power-management protocol based on sleep/wakeup schedules. The protocol is well suited for data collection applications in which sensors have to periodically report to a sink. It staggers the schedules of the nodes, in order to minimize the delay. One major advantage of this protocol is that the schedules are automatically adapted based on the network congestion and on the application traffic demand, so that the network can operate efficiently and completely unattended even in very dynamic conditions. Simulation results show that our power management protocol effectively reacts to traffic and topology variations, without scarifying performance in terms of energy consumption, delivery ratio and delay. Furthermore it achieves lower energy consumption, collision ratio and delay than commonly adopted fixed sleep/wakeup schemes.